Does Domestic Investment Produce Economic Growth in Canada: Empirical Analysis Based on Correlation, Cointegration and Causality

(1)

Munich Personal RePEc Archive

Does Domestic Investment Produce

Economic Growth in Canada: Empirical Analysis Based on Correlation,

Cointegration and Causality

Bakari, Sayef

Department of Economics Science, LIEI, Faculty of Economic Sciences and Management of Tunis (FSEGT), University Of Tunis El Manar, Tunisia.

8 December 2016

Online at https://mpra.ub.uni-muenchen.de/75966/

MPRA Paper No. 75966, posted 04 Jan 2017 08:46 UTC

(2)

1

Does Domestic Investment Produce Economic Growth in Canada: Empirical Analysis Based on Correlation,

Cointegration and Causality

Sayef Bakari

PhD Student, Department of Economics Science, LIEI, Faculty of Economic Sciences and Management of Tunis (FSEGT), University Of Tunis El Manar, Tunisia.

Email: bakari.sayef@yahoo.fr Abstract

This paper investigates the relationship between domestic investment and economic growth in Canada. In order to achieve this purpose, annual data for the periods between 1990 and 2015 was tested by using Correlation analysis, Johansen co-integration analysis of Vector Error Correction Model and the Granger-Causality tests. According to the result of the analysis, it was determined that there is no relationship between the four variables in the long run term, however, there is a weak relationship between domestic investment and economic growth in the short run term. On the other hand, the results of the Granger Causality test show that there is no causal relationship between domestic investment and economic growth. The result provide that domestic investment affects economic growth on the short run term, however the domestic investment does not cause economic growth in Canada

KEYWORDS: Domestic Investment, Economic Growth, Canada, Correlation, Cointegration, VECM and Causality.

JEL Classification: C13, E22.

I. Introduction

Investment is considered the cornerstone of economic and social development as the primary motivation for growth through increased domestic product and provides the raw materials extra complement national savings and resources investable within each country. Respect of domestic investment at the level of the national economy, capital spending on new projects in the sectors of public utilities and infrastructure such as incision main and branch roads projects and extensions of water and sewerage connections and create urban plans and construction projects, housing and extensions of electricity and power generation, as well as

(3)

2

social development in the areas of education, health and communication projects, projects as well to projects that relate to economic activity for the production of goods and services in the production and service sectors such as industry, agriculture, housing, health, education and tourism. Canada is the world's tenth largest economy in 2014 with a gross domestic product of

$ 1,887 billion. The Canadian economy is strongly linked to the US economy, due to geographic proximity and commercial treaties. The latest value for Gross fixed capital formation (current US$) in Canada was $422,415,000,000 as of 2014. Over the past 54 years, the value for this indicator has fluctuated between $443,961,000,000 in 2012 and

$8,631,470,000 in 1961. Also, Canadians prefer to invest locally. For example, 76% of their total investment portfolio (equities, bonds and mutual fund units) is invested in Canada. In particular, this work tries to empirically find an answer for the question of whether there is a nexus between domestic investment and economic growth in Canada, to achieve this objective the paper is structured as follows. In section 2, we present the review literature concerning the nexus between trade and economic growth. Secondly, we discuss the Methodology Model Specification and data used in this study in Section 3. Thirdly, Section 4 and section 5 presents the empirical results as well as the analysis of the findings. Finally, Section 6 is dedicated to our conclusion.

II. Literature Review

Several empirical studies which investigated the relationship between domestic investment and economic growth found that, fixed capital formation determine the rate of future economic growth. These studies include:

Blomstorm and al (1994) found that, capital formation does not cause economic growth, instead the causal direction flows from economic growth to capital formation.

Ghali and Al-Mutawa (1999) investigated the causality relationship between gross fixed capital formation and economic growth using VAR. The results varied significantly across countries; for the case of Japan and UK there was feedback causality relationship, whilst in USA and France there was unidirectional causality from fixed investment to economic growth. However, for the case of Canada, Germany and Italy the economic growth rate caused fixed capital formation.

Sumei Tang, E. A. Selvanathan and S. Selvanathan (2008) investigates the causal link between foreign direct investment (FDI), domestic investment and economic growth in China

(4)

3

for the period 1988–2003 using a multivariate VAR system with error correction model (ECM). The results show that while there is a bi-directional causality between domestic investment and economic growth, there is only single-directional causality from FDI to domestic investment and to economic growth.

Ghazali (2010) identified the causal relationship between private domestic investment and economic growth (GDP) in Pakistan over the period 1981 to 2008. He discovered that there is a bi-directional causality between private domestic investment and economic growth.

Adhikary (2011) found that, capital formation has long run relationship with export and import in Bangladesh. On the other hand, the study found long run causality relationship flows from trade, capital formation and FDI to economic growth. In this way the study concluded that, capital formation has long run relationship and cause economic growth.

Bakare (2011) studied the relationship between capital formation and growth rate with respect to Nigerian Economy using Harrod–Domar model. Using error correction mechanism, the study found out a positive long-run relationship between capital formation and economic growth in Nigeria

Tan and Tang (2011) investigated the dynamic relationship between private domestic investment (PDI), the user cost of capital and economic growth in Malaysia over the period of 1970 to 2009. His result shows that PDI, the user cost of capital, and economic growth are cointegrated in Malaysia. The Granger causality test shows that there is a unidirectional causality running from PDI to economic growth and from PDI to the user cost of capital in the long run.

Debi Prasad Bal, Devi Prasad Dash and Bibhudutta Subhasish (2016) examine the impact of capital formation on economic growth in India covering the period from 1970 to 2012.The error correction (ECM) model shows that the capital formation positively affects the economic growth in the short run. It is recommended that government increases the level of capital formation in order to achieve a higher level of economic growth.

III. Data, methodology and model specification:

1. The Data:

The analysis used in this study cover annual time series of 1990 to 2015 or 26 observations which should be sufficient to capture the short run and long run correlation between Export, Import, Fixed Formation Capital and economic growth in the model. The data set consists of

(5)

4

observation for GDP, exports of goods and services (current US$), imports of goods and services (current US$) and Fixed Formation Capital (current US$). All data set are taken from World Development Indicators 2016.

2. Methodology

We will use the most appropriate method which consists firstly of determining the degree of integration of each variable. If the variables are all integrated in level, we apply an estimate based on a linear regression. On the other hand, if the variables are all integrated into the first difference, our estimates are based on an estimate of the VAR model. When the variables are integrated in the first difference we will examine and determine the cointegration between the variables, if the cointegration test indicates the absence of cointegration relation, we will use the model VAR. If the cointegration test indicates the presence of a cointegration relation between the different variables studied, the model VECM will be used.

3. Model specification:

Early empirical formulations tried to capture the causal link between domestic investment and GDP growth by incorporating exports into the aggregate production function (Balassa, 1978;

Masoud Albiman Md and Suleiman NN, 2016). The augmented production function including domestic investment, exports and imports is expressed as:

𝑮𝑫𝑷𝒕 = 𝒇(𝒆𝒙𝒑𝒐𝒓𝒕𝒔, 𝒊𝒎𝒑𝒐𝒓𝒕𝒔, 𝒄𝒂𝒑𝒊𝒕𝒂𝒍) (1)

The function can also be represented in a log-linear econometric format thus:

𝐥𝐨𝐠⁡(𝑮𝑫𝑷)𝒕 = 𝜷𝟎+ 𝜷𝟏𝐥𝐨𝐠⁡(𝒆𝒙𝒑𝒐𝒓𝒕𝒔)𝒕+ 𝜷𝟐𝐥𝐨𝐠⁡(𝒊𝒎𝒑𝒐𝒓𝒕𝒔)𝒕+ 𝜷𝟑𝐥𝐨𝐠⁡(𝒄𝒂𝒑𝒊𝒕𝒂𝒍)𝒕+ 𝜺𝒕 (2) Where:

- 𝛽₀ : The constant term.

- 𝛽1: coefficient of variable (exports) - 𝛽₂: coefficient of variables (imports) - 𝛽₃: coefficient of variable (capital) - 𝑡: The time trend.

- 𝜀: The random error term assumed to be normally, identically and independently distributed.

(6)

5 IV. Empirical Analysis

A- Empirical Results 1. Statistic Descriptive

Table 1: Statistic Descriptive

LOG(GDP) LOG(EXPORTS) LOG(IMPORTS) LOG(CAPITAL)

Mean 27.59252 26.50076 26.46350 26.04641

Median 27.43552 26.51783 26.39529 25.82931

Maximum 28.23939 27.05952 27.09850 26.82789

Minimum 27.08140 25.72410 25.71952 25.40569

Std. Dev. 0.447394 0.449994 0.467829 0.530072

Skewness 0.239267 -0.374991 -0.086874 0.254517

Kurtosis 1.368956 1.862324 1.671351 1.382617

Jarque-Bera 3.130075 2.011511 1.945120 3.114630

Probability 0.209080 0.365768 0.378114 0.210701

Sum 717.4056 689.0197 688.0510 677.2065

Sum Sq. Dev. 5.004042 5.062369 5.471600 7.024409

Observations 26 26 26 26

2. The Correlation Test

Table 2: Pearson correlation coefficient value

LOG(GDP) LOG(CAPITAL) LOG(EXPORTS) LOG(IMPORTS)

LOG(GDP) 1 0.9979023795966825 0.9346471798342778 0.9699481014528804 LOG(CAPITAL) 0.9979023795966825 1 0.9240457539770799 0.9627610435633275 LOG(EXPORTS) 0.9346471798342778 0.9240457539770799 1 0.988828116726709 LOG(IMPORTS) 0.9699481014528804 0.9627610435633275 0.988828116726709 1

(7)

6 3. Test for unit root

Table 3: Test for unit root of Log (GDP) Test for unit root in level

LOG(GDP) Exogenous: Constant Exogenous: Constant, Linear

Trend Augmented Dickey-Fuller test

statistic

t-Statistic Prob.* t-Statistic Prob.*

-0.443071 0.8867 -1.533221 0.7901

Test critical values:

1% level -3.724070 -4.374307

5% level -2.986225 -3.603202

10% level -2.632604 -3.238054

Test for unit root in first difference

statistic

-2.911310 0.0588 -2.738967

0.2312 Test critical

values:

1% level -3.737853 -4.394309

5% level -2.991878 -3.612199

10% level -2.635542 -3.243079

Test for unit root in second difference

statistic

-5.230467 0.0004 -5.653833 0.0008

1% level -3.769597 -4.440739

5% level -3.004861 -3.632896

10% level -2.642242 -3.254671

Graph 1: Evolution of Log (GDP)

27. 0 27. 2 27. 4 27. 6 27. 8 28. 0 28. 2 28. 4

90 92 94 96 98 00 02 04 06 08 10 12 14

LOG(GDP)

(8)

7

Table 4: Test for unit root of Log (Exports) Test for unit root in level

LOG(EXPORTS) Exogenous: Constant Exogenous: Constant,

Linear Trend

Augmented Dickey-Fuller test statistic t-Statistic Prob.* t-Statistic Prob.*

-1.572330 0.4814 -1.510860 0.7984

1% level -3.724070 -4.374307

5% level -2.986225 -3.603202

10% level -2.632604 -3.238054

Linear Trend

-4.584968 0.0014 -4.884366 0.0035

1% level -3.737853 -4.394309

5% level -2.991878 -3.612199

10% level -2.635542 -3.243079

Linear Trend

-5.755822 0.0001 -5.714225 0.0007

1% level -3.769597 -4.440739

5% level -3.004861 -3.632896

10% level -2.642242 -3.254671

Graph 2: Evolution of Log (Exports)

25.6 25.8 26.0 26.2 26.4 26.6 26.8 27.0 27.2

90 92 94 96 98 00 02 04 06 08 10 12 14

LOG(EXPORTS)

(9)

8

Table 5: Test for unit root of Log (Imports) Test for unit root in level

LOG(IMPORTS) Exogenous: Constant Exogenous: Constant, Linear

Trend

-1.211660 0.6529 -1.907362 0.6208

1% level -3.724070 -4.374307

5% level -2.986225 -3.603202

10% level -2.632604 -3.238054

Trend

-4.562211 0.0015 -4.646242 0.0058

1% level -3.737853 -4.394309

5% level -2.991878 -3.612199

10% level -2.635542 -3.243079

Trend

-7.424952 0.0000 -7.366850 0.0000

1% level -3.752946 -4.416345

5% level -2.998064 -3.622033

10% level -2.638752 -3.248592

Graph 3: Evolution of Log (Imports)

25. 6 25. 8 26. 0 26. 2 26. 4 26. 6 26. 8 27. 0 27. 2

90 92 94 96 98 00 02 04 06 08 10 12 14

LOG(IMPORTS)

(10)

9

Table 6: Test for unit root of Log (Domestic investment) Test for unit root in level

LOG(DOMESTIC INVESTMENT) Exogenous: Constant

Exogenous: Constant, Linear Trend

-0.253327 0.9188 -2.049736 0.5472

Test critical values: 1% level -3.724070 -4.374307

5% level -2.986225 -3.603202

10% level -2.632604 -3.238054

-3.045024 0.0449 -2.813204 0.2063

5% level -2.991878 -3.612199

10% level -2.635542 -3.243079

-6.827934 0.0000 -7.079465 0.0000

5% level -2.998064 -3.622033

10% level -2.638752 -3.248592

Graph 4: Evolution of Log (Domestic investment)

25. 2 25. 4 25. 6 25. 8 26. 0 26. 2 26. 4 26. 6 26. 8 27. 0

90 92 94 96 98 00 02 04 06 08 10 12 14

LOG(CAPITAL)

(11)

10 4. Lag order selection criteria

Table 7: VAR Lag Order Selection Criteria VAR Lag Order Selection Criteria

Endogenous variables: LOG(GDP) LOG(EXPORTS) LOG(IMPORTS) LOG(CAPITAL) Exogenous variables: C

Sample: 1990 2015 Included observations: 23

Lag LogL LR FPE AIC SC HQ

0 109.2926 NA 1.24e-09 -9.155882 -8.958405 -9.106217

1 190.5927 127.2523 4.36e-12 -14.83415 -13.84676 -14.58582 2 215.9462 30.86514* 2.24e-12* -15.64750 -13.87020* -15.20051*

3 231.9584 13.92367 3.43e-12 -15.64856* -13.08136 -15.00292 * indicates lag order selected by the criterion

LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error

AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion

5. Cointegration analysis

Table 8: Cointegration Test Sample (adjusted): 1993 2015

Included observations: 23 after adjustments Trend assumption: Linear deterministic trend

Series: LOG(GDP) LOG(EXPORTS) LOG(IMPORTS) LOG(CAPITAL) Lags interval (in first differences): 1 to 2

Unrestricted Cointegration Rank Test (Trace) Hypothesized No. of

CE(s) Eigenvalue Trace Statistic Critical Value 0.05 Prob.**

None * 0.843710 76.42521 47.85613 0.0000

At most 1 * 0.576441 33.73619 29.79707 0.0167

(12)

11

At most 2 0.329718 13.97774 15.49471 0.0836

At most 3 * 0.187525 4.776421 3.841466 0.0288

Trace test indicates 2 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values 1 Cointegrating

Equation(s): Log likelihood 215.0903

Normalized cointegrating coefficients (standard error in parentheses)

LOG(GDP) LOG(EXPORTS) LOG(IMPORTS) LOG(CAPITAL)

1.000000 0.180846 0.054395 -0.999679

(0.08348) (0.09779) (0.03200)

6. Estimation of VECM

a) Estimation of The Long Run Equation

Table 9: Estimation of Long Run Equation by Using Least Squares Dependent Variable: D(LOG(GDP))

Method: Least Squares (Gauss-Newton / Marquardt steps) Sample (adjusted): 1993 2015

Included observations: 23 after adjustments

Coefficient Std. Error t-Statistic Prob.

C(1) 0.887376 1.045781 0.848529 0.4127

C(2) -1.169103 0.387541 -3.016721 0.0107

C(3) 1.134698 0.948253 1.196620 0.2546

C(4) -1.189003 0.620695 -1.915599 0.0795

C(5) 0.159053 0.711334 0.223598 0.8268

C(6) 0.136578 0.728304 0.187529 0.8544

C(7) -1.964493 1.147575 -1.711864 0.1126

C(8) -0.494156 1.070382 -0.461663 0.6526

C(9) 1.126319 0.652770 1.725446 0.1101

C(10) 1.383214 0.659001 2.098956 0.0577

C(11) 0.040730 0.024043 1.694052 0.1160

b) Estimation of The Short Run Equation

(13)

12

Table 10: Estimation of Short Run Equation by Using Wald Test Wald Test: Exports

Equation: Untitled

Test Statistic Value Df Probability

F-statistic 0.037093 (2, 12) 0.9637

Chi-square 0.074186 2 0.9636

Null Hypothesis: C(6)=C(5)=0 Null Hypothesis Summary:

Normalized Restriction (= 0) Value Std. Err.

C(6) 0.136578 0.728304

C(5) 0.159053 0.711334

Restrictions are linear in coefficients.

Wald Test: Imports

Test Statistic Value df Probability

F-statistic 1.546781 (2, 12) 0.2525

Chi-square 3.093561 2 0.2129

C(8) -0.494156 1.070382

C(7) -1.964493 1.147575

Wald Test: Domestic Investment Equation: Untitled

Test Statistic Value df Probability

F-statistic 2.768587 (2, 12) 0.1026

Chi-square 5.537174 2 0.0628

C(10) 1.383214 0.659001

C(9) 1.126319 0.652770

Restrictions are linear in coefficients.

(14)

13 7. Checking the quality of the model a) Serial Correlation LM Test

Table 11: Serial Correlation LM Test

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 0.224380 Prob. F(2,10) 0.8029

Obs*R-squared 0.987820 Prob. Chi-Square(2) 0.6102

b) Heteroskedasticity Tests

Table 12: Heteroskedasticity Tests Heteroskedasticity Test: Breusch-Pagan-Godfrey

Scaled explained SS 1.199706 Prob. Chi-Square(12) 1.0000

Heteroskedasticity Test: ARCH

Heteroskedasticity Test: Harvey

Heteroskedasticity Test: Glejser

c) Test of Normality

(15)

14

Graph 5: Test of Normality

d) VAR Stability

 Test CUSUM

Graph 6: Test CUSUM

 Test CUSUM of Squares

Graph 7: Test CUSUM of Squares

0 1 2 3 4 5 6 7 8

-0.10 -0.05 0.00 0.05

Series: Residuals Sample 1993 2015 Observations 23

Mean 8.64e-16 Median -0.001779 Maximum 0.056324 Minimum -0.084613 Std. Dev. 0.045234 Skewness -0.488234 Kurtosis 2.091522 Jarque-Bera 1.704706 Probability 0.426410

-12 -8 -4 0 4 8 12

04 05 06 07 08 09 10 11 12 13 14 15

CUS UM 5% S ignific anc e

-0.4 0.0 0.4 0.8 1.2 1.6

04 05 06 07 08 09 10 11 12 13 14 15

CUS UM of S quares 5% S ignific anc e

(16)

15 8. Granger Causality Tests

Table 9: Ganger Causality Tests Pairwise Granger Causality Tests

Sample: 1990 2015 Lags: 2

Null Hypothesis: F-Statistic Prob.

LOG(EXPORTS) does not Granger Cause LOG(GDP) 5.35016 0.0144

LOG(GDP) does not Granger Cause LOG(EXPORTS) 1.64031 0.2202

LOG(IMPORTS) does not Granger Cause LOG(GDP) 3.57611 0.0481

LOG(GDP) does not Granger Cause LOG(IMPORTS) 1.83294 0.1871

LOG(CAPITAL) does not Granger Cause LOG(GDP) 0.47499 0.6291

LOG(GDP) does not Granger Cause LOG(CAPITAL) 2.00253 0.1625

LOG(IMPORTS) does not Granger Cause LOG(EXPORTS) 0.52465 0.6001

LOG(EXPORTS) does not Granger Cause LOG(IMPORTS) 1.16225 0.3340

LOG(CAPITAL) does not Granger Cause LOG(EXPORTS) 1.72228 0.2054

LOG(EXPORTS) does not Granger Cause LOG(CAPITAL) 3.50380 0.0507

LOG(CAPITAL) does not Granger Cause LOG(IMPORTS) 2.80716 0.0855

LOG(IMPORTS) does not Granger Cause LOG(CAPITAL) 3.03324 0.0719

B- Interpretation

Table 1 presents the descriptive statistics of our database used in our research work by presenting and describing the mean, median, maximum and minimum of each variable.

Otherwise, and according to the application of descriptive statistics, one notices the existence of a heteroskedasticity between the different variables. The results of the test of correlation (Table2) show the relationship between the variables is positively correlated. According to the correlation matrix of the variables, it is found that the dependent variable (GDP) and the independent variable (exports) are positively correlated with a correlation coefficient equal to (0.9346471798342778). Thus, if exports increase by 1%, gross domestic product (GDP) increases by 0.9346471798342778%. Otherwise, the dependent variable (GDP) and the independent variable (imports) are positively correlated with a correlation coefficient equal to (0.9699481014528804). Thus, if imports increase by 1%, the gross domestic product (GDP) increases by 0.9699481014528804%. Finally, we notice that the dependent variable (GDP)

(17)

16

and the independent variable (domestic investment) are positively correlated with a correlation coefficient equal to (0.9979023795966825), which is meaning that if domestics’

investments increase by 1%, the gross domestic product (GDP) increase by 0.9979023795966825%. In order to evaluate the degree of integration of each variable, we use Augmented Dickey-Fuller (ADF) test (Table, 3, 4, 5 and 6). The results show that all the variables are not stationary in level, for the first difference we note that the variable log (PIB) is not stationary, if we pass to the second difference we remark that all variables becomes stationary. This forces us to go directly from verifying if there is a co-integration of the variables. Therefore, and according to these interpretations of the stationary of the variables, we will use the VAR model. For this reason we have applied the test of cointegration and this is done after the choice of number of delay existing in our model. It is clear that LR, FPE, AIC, SC, HQ and HQ statistics are chosen lag 2 for each endogenous variable in their autoregressive and distributed lag structures in the estimable VAR model. Therefore, lag of 2 is used for estimation purpose (Table7). The results of the cointegration test (Table 8) indicates 2 cointegrating equations at the 0.05 level, and provide the existence of long run equation between GDP, exports, imports and domestic investment. According to this long run equation a 1% increase in exports leads to a decrease of 0.180846 % on GDP. On the other hand, a 1% increase in imports leads to a decrease of 0.054396% on GDP. Also, we observe that a 1% increase in domestic investment leads to an increase of 0.999679% on GDP.

Otherwise, the four variables are cointegrated, which obliges us to use the VECM model to test the significance of this model. The results of the correction error model show that there is no relationship between the four variables in the long run term (Table 9). In the otherwise, the correction error model shows that there is a relationship between domestic investment and GDP in the short run term (Table 10). To check the quality of our model and to ensure the robustness of our estimate, there is a set of tests and indicators that designates and affirms that our work is acceptable or not. Among these tests are: Serial Correlation (Table 11), Heteroskedasticity tests (Table 12), test of Normality (Graph 5) and the Var stability (Graph 6 and Graph 7). Finally, the results of Granger Causality Tests show that there is no relationship of causality between investment domestic and GDP.

V. Conclusion

The aim of this study was to explain the nexus between domestic investment and economic growth in Canada during the period 1990-2015. The Pearson Correlation Coefficient Value, Co-integration, Vector Error Correction Model and Granger’s Causality tests are applied to

(18)

17

investigate the relationship between these three variables. The unit root properties of the data were examined using the Augmented Dickey Fuller test (ADF) after that the cointegration and causality tests were conducted. The empirical results show that there is no relationship between the four variables in the long run term, however, there is a weak relationship between domestic investment and economic growth in the short run term. On the other hand, and according to the results of the Granger Causality test shows that there is no causal relationship between domestic investment and economic growth. These results provide evidence that domestic investment, thus, is seen as the source of economic growth in Canada on the short term, however it seen also that growth in Canada was propelled by growth-led trade strategy.

Reference:

Adhikary BK (2011) FDI, trade openness, capital formation and economic growth in Bangladesh: A linkage analysis. International Journal of Business and Management 6: 17-27.

Bakare, A.S. (2011). The determinants and roles of capital flight in the growth process of Nigerian economy: Vector autoregressive model approach. British Journal of Management and Economics, 1(2), 100–113.

Balassa, B., (1978) ‘‘Export and Economic Growth: Further Evidence’’, Journal of Development Economics, 5, 181 – 189.

Blomstrom M, Lipsey R, Zejan M (1994). What explains the developing countries growth?

Oxford University Press, UK.

Debi Prasad Bal, Devi Prasad Dash and Bibhudutta Subhasish (2016). The Effects of Capital Formation on Economic Growth in India: Evidence from ARDL-bound Testing Approach.

Global Business Review. 17(6) 1388–1400.

Dickey, D. A. & W. A. Fuller (1979), “Distribution of Estimators of Autoregressive Time Series with a Unit Root,” Journal of the American Statistical Association, 74, 427-31.

(19)

18

Dickey, D. A. & W. A. Fuller (1981) “Likelihood ratio Statistics for autoregressive time series with a unit root,” Econometrica, 49(4):1057-72.

Engle, R. F. & Granger C. W. (1987), “Cointegration and Error Correction: Representation, Estimation and Testing,” Econometrica, 55, 251-276.

Ghali KH, Al-Mutawa A (1999). The intertemporal causal dynamics between fixed capital formation and economic growth in the group of seven countries. International Economic Journal 13: 31-37.

Ghazali, A. (2010). Analyzing the Relationship between Foreign Direct Investment, Domestic investment and economic growth for Pakistan. International Research Journal of Finance and Economics, 47, 123-131.

Imoisi Anthony Ilegbinosa, Abuo Micheal and Sogules Ipalibo Watson (2015). Domestic Investment and Economic Growth in Nigeria from 1970-2013: An Econometric Analysis Canadian Social Science Vol. 11, No. 6, 2015, pp. 70-79.

Johansen, S. (1988), “Statistical Analysis of Cointegration Vectors,” Journal of Economic Dynamics and Control, 12, 231-54.

Sumei Tang, E. A. Selvanathan and S. Selvanathan (2008).Foreign Direct Investment, Domestic Investment and Economic Growth in China: A Time Series Analysis. The World Economy. Volume 31, Issue 10 October 2008. Pages 1292–1309.

Tan, B. W., & Tang, C. F (2011). The dynamic relationship between private domestic investment, the user cost of capital and economic growth in Malaysia. MPRA Paper No.

27964.